GAN-based image attribute editing firstly leverages GAN Inversion to project real images into the latent space of GAN and then manipulates corresponding latent codes. Recent inversion methods mainly utilize additional high-bit features to improve image details preservation, as low-bit codes cannot faithfully reconstruct source images, leading to the loss of details. However, during editing, existing works fail to accurately complement the lost details and suffer from poor editability. The main reason is they inject all the lost details indiscriminately at one time, which inherently induces the position and quantity of details to overfit source images, resulting in inconsistent content and artifacts in edited images. This work argues that details should be gradually injected into both the reconstruction and editing process in a multi-stage coarse-to-fine manner for better detail preservation and high editability. Therefore, a novel dual-stream framework is proposed to accurately complement details at each stage. The Reconstruction Stream is employed to embed coarse-to-fine lost details into residual features and then adaptively add them to the GAN generator. In the Editing Stream, residual features are accurately aligned by our Selective Attention mechanism and then injected into the editing process in a multi-stage manner. Extensive experiments have shown the superiority of our framework in both reconstruction accuracy and editing quality compared with existing methods.
Sequential Recommendation (SR) captures users' dynamic preferences by modeling how users transit among items. However, SR models that utilize only single type of behavior interaction data encounter performance degradation when the sequences are short. To tackle this problem, we focus on Multi-Behavior Sequential Recommendation (MBSR) in this paper, which aims to leverage time-evolving heterogeneous behavioral dependencies for better exploring users' potential intents on the target behavior. Solving MBSR is challenging. On the one hand, users exhibit diverse multi-behavior patterns due to personal characteristics. On the other hand, there exists comprehensive co-influence between behavior correlations and item collaborations, the intensity of which is deeply affected by temporal factors. To tackle these challenges, we propose a Personalized Behavior-Aware Transformer framework (PBAT) for MBSR problem, which models personalized patterns and multifaceted sequential collaborations in a novel way to boost recommendation performance. First, PBAT develops a personalized behavior pattern generator in the representation layer, which extracts dynamic and discriminative behavior patterns for sequential learning. Second, PBAT reforms the self-attention layer with a behavior-aware collaboration extractor, which introduces a fused behavior-aware attention mechanism for incorporating both behavioral and temporal impacts into collaborative transitions. We conduct experiments on three benchmark datasets and the results demonstrate the effectiveness and interpretability of our framework. Our implementation code is released at https://github.com/TiliaceaeSU/PBAT.
In order to circumvent statistical and computational hardness results in sequential decision-making, recent work has considered smoothed online learning, where the distribution of data at each time is assumed to have bounded likeliehood ratio with respect to a base measure when conditioned on the history. While previous works have demonstrated the benefits of smoothness, they have either assumed that the base measure is known to the learner or have presented computationally inefficient algorithms applying only in special cases. This work investigates the more general setting where the base measure is \emph{unknown} to the learner, focusing in particular on the performance of Empirical Risk Minimization (ERM) with square loss when the data are well-specified and smooth. We show that in this setting, ERM is able to achieve sublinear error whenever a class is learnable with iid data; in particular, ERM achieves error scaling as $\tilde O( \sqrt{\mathrm{comp}(\mathcal F)\cdot T} )$, where $\mathrm{comp}(\mathcal F)$ is the statistical complexity of learning $\mathcal F$ with iid data. In so doing, we prove a novel norm comparison bound for smoothed data that comprises the first sharp norm comparison for dependent data applying to arbitrary, nonlinear function classes. We complement these results with a lower bound indicating that our analysis of ERM is essentially tight, establishing a separation in the performance of ERM between smoothed and iid data.
Wi-Fi fingerprinting has emerged as the most popular approach to indoor localization. The use of ML algorithms has greatly improved the localization performance of Wi-Fi fingerprinting, but its success depends on the availability of fingerprint databases composed of a large number of RSSIs, the MAC addresses of access points, and the other measurement information. However, most fingerprint databases do not reflect well the time varying nature of electromagnetic interferences in complicated modern indoor environment. This could result in significant changes in statistical characteristics of training/validation and testing datasets, which are often constructed at different times, and even the characteristics of the testing datasets could be different from those of the data submitted by users during the operation of localization systems after their deployment. In this paper, we consider the implications of time-varying Wi-Fi fingerprints on indoor localization from a data-centric point of view and discuss the differences between static and dynamic databases. As a case study, we have constructed a dynamic database covering three floors of the IR building of XJTLU based on RSSI measurements, over 44 days, and investigated the differences between static and dynamic databases in terms of statistical characteristics and localization performance. The analyses based on variance calculations and Isolation Forest show the temporal shifts in RSSIs, which result in a noticeable trend of the increase in the localization error of a Gaussian process regression model with the maximum error of 6.65 m after 14 days of training without model adjustments. The results of the case study with the XJTLU dynamic database clearly demonstrate the limitations of static databases and the importance of the creation and adoption of dynamic databases for future indoor localization research and real-world deployment.
We study the integration of machine learning advice into the design of skip lists to improve upon traditional data structure design. Given access to a possibly erroneous oracle that outputs estimated fractional frequencies for search queries on a set of items, we construct a skip list that provably provides the optimal expected search time, within nearly a factor of two. In fact, our learning-augmented skip list is still optimal up to a constant factor, even if the oracle is only accurate within a constant factor. We show that if the search queries follow the ubiquitous Zipfian distribution, then the expected search time for an item by our skip list is only a constant, independent of the total number $n$ of items, i.e., $\mathcal{O}(1)$, whereas a traditional skip list will have an expected search time of $\mathcal{O}(\log n)$. We also demonstrate robustness by showing that our data structure achieves an expected search time that is within a constant factor of an oblivious skip list construction even when the predictions are arbitrarily incorrect. Finally, we empirically show that our learning-augmented skip list outperforms traditional skip lists on both synthetic and real-world datasets.
The vision of pervasive artificial intelligence (AI) services can be realized by training an AI model on time using real-time data collected by internet of things (IoT) devices. To this end, IoT devices require offloading their data to an edge server in proximity. However, transmitting high-dimensional and voluminous data from energy-constrained IoT devices poses a significant challenge. To address this limitation, we propose a novel offloading architecture, called joint data deepening-and-prefetching (JD2P), which is feature-by-feature offloading comprising two key techniques. The first one is data deepening, where each data sample's features are sequentially offloaded in the order of importance determined by the data embedding technique such as principle component analysis (PCA). Offloading is terminated once the already transmitted features are sufficient for accurate data classification, resulting in a reduction in the amount of transmitted data. The criteria to offload data are derived for binary and multi-class classifiers, which are designed based on support vector machine (SVM) and deep neural network (DNN), respectively. The second one is data prefetching, where some features potentially required in the future are offloaded in advance, thus achieving high efficiency via precise prediction and parameter optimization. We evaluate the effectiveness of JD2P through experiments using the MNIST dataset, and the results demonstrate its significant reduction in expected energy consumption compared to several benchmarks without degrading learning accuracy.
Natural Language Processing (NLP) technologies have revolutionized the way we interact with information systems, with a significant focus on converting natural language queries into formal query languages such as SQL. However, less emphasis has been placed on the Corpus Query Language (CQL), a critical tool for linguistic research and detailed analysis within text corpora. The manual construction of CQL queries is a complex and time-intensive task that requires a great deal of expertise, which presents a notable challenge for both researchers and practitioners. This paper presents the first text-to-CQL task that aims to automate the translation of natural language into CQL. We present a comprehensive framework for this task, including a specifically curated large-scale dataset and methodologies leveraging large language models (LLMs) for effective text-to-CQL task. In addition, we established advanced evaluation metrics to assess the syntactic and semantic accuracy of the generated queries. We created innovative LLM-based conversion approaches and detailed experiments. The results demonstrate the efficacy of our methods and provide insights into the complexities of text-to-CQL task.
Recent advancements in generative large language models (LLMs) have significantly boosted the performance in natural language processing tasks. However, their efficiency is hampered by the inherent limitations in autoregressive token generation. While parallel decoding with token tree verification, e.g., Medusa, has been proposed to improve decoding parallelism and efficiency, it often struggles with maintaining contextual relationships due to its independent token prediction approach and incurs significant verification overhead, especially with large tree sizes and batch processing. In this paper, we propose ProPD, an efficient LLM parallel decoding framework based on dynamic token tree pruning and generation. ProPD features an advanced early pruning mechanism to efficiently eliminate unpromising token sequences to improve verification efficiency. Additionally, it introduces a dynamic token tree generation algorithm to balance the computation and parallelism of the verification phase in real-time and maximize the overall efficiency across different batch sizes, sequence lengths, and tasks, etc. We verify ProPD across a diverse set of datasets, LLMs, and batch sizes and demonstrate ProPD consistently outperforms existing decoding algorithms by 1.1-3.2x.
Imbalanced data, characterized by an unequal distribution of data points across different clusters, poses a challenge for traditional hard and fuzzy clustering algorithms, such as hard K-means (HKM, or Lloyd's algorithm) and fuzzy K-means (FKM, or Bezdek's algorithm). This paper introduces equilibrium K-means (EKM), a novel and simple K-means-type algorithm that alternates between just two steps, yielding significantly improved clustering results for imbalanced data by reducing the tendency of centroids to crowd together in the center of large clusters. We also present a unifying perspective for HKM, FKM, and EKM, showing they are essentially gradient descent algorithms with an explicit relationship to Newton's method. EKM has the same time and space complexity as FKM but offers a clearer physical meaning for its membership definition. We illustrate the performance of EKM on two synthetic and ten real datasets, comparing it to various clustering algorithms, including HKM, FKM, maximum-entropy fuzzy clustering, two FKM variations designed for imbalanced data, and the Gaussian mixture model. The results demonstrate that EKM performs competitively on balanced data while significantly outperforming other techniques on imbalanced data. For high-dimensional data clustering, we demonstrate that a more discriminative representation can be obtained by mapping high-dimensional data via deep neural networks into a low-dimensional, EKM-friendly space. Deep clustering with EKM improves clustering accuracy by 35% on an imbalanced dataset derived from MNIST compared to deep clustering based on HKM.
In this paper, we tackle the problem of computing a sequence of rankings with the guarantee of the Pareto-optimal balance between (1) maximizing the utility of the consumers and (2) minimizing unfairness between producers of the items. Such a multi-objective optimization problem is typically solved using a combination of a scalarization method and linear programming on bi-stochastic matrices, representing the distribution of possible rankings of items. However, the above-mentioned approach relies on Birkhoff-von Neumann (BvN) decomposition, of which the computational complexity is $\mathcal{O}(n^5)$ with $n$ being the number of items, making it impractical for large-scale systems. To address this drawback, we introduce a novel approach to the above problem by using the Expohedron - a permutahedron whose points represent all achievable exposures of items. On the Expohedron, we profile the Pareto curve which captures the trade-off between group fairness and user utility by identifying a finite number of Pareto optimal solutions. We further propose an efficient method by relaxing our optimization problem on the Expohedron's circumscribed $n$-sphere, which significantly improve the running time. Moreover, the approximate Pareto curve is asymptotically close to the real Pareto optimal curve as the number of substantial solutions increases. Our methods are applicable with different ranking merits that are non-decreasing functions of item relevance. The effectiveness of our methods are validated through experiments on both synthetic and real-world datasets.